PKPD-Modelling of QT Prolongation Following Deliberate Self-Poisonings with Citalopram

Transcription

1 PKPD-Modelling of QT Prolongation Following Deliberate Self-Poisonings with Citalopram Lena Friberg 1, Geoffrey Isbister 2, Peter Hackett 3 and Stephen Duffull 1 1 School of Pharmacy, University of Queensland, Australia 2 Clinical Toxicology and Pharmacology, Newcastle Mater Hospital and a Tropical Toxinology Unit, Charles Darwin University, Australia 3 Clinical Pharmacology and Toxicology, Western Australian Centre for Pathology and Medical Research, Australia

2 PKPD studies in clinical toxicology No controlled studies Uncertainty in dosing history Co-ingested drugs Decontamination procedures Activated charcoal Vomiting Sparse sampling Few samples in absorption phase?

3 Analysing PK and PD data from overdoses Population PKPD analysis Fully Bayesian methodology Prior information Posterior distribution Prior distribution Likelihood distribution p( θ y) p( θ ) p( yθ ) Uncertainty in dose and time

4 Citalopram in overdose Antidepressant SSRI QT prolongation in a larger frequency than other SSRIs (Isbister et al, 2004) Documented cases of Torsade de Pointes (TdP) (Tarabar et al, 2003; Meuleman et al, 2001) Several fatal cases described (Öström m et al, 1996; Jonasson and Saldeen,, 2002)

5 QT-RR R RR interval ~ 1 / Heart Rate QT interval T P Q S Fossa et al. J Pharmacol Exp Ther, 2005

6 Aim To develop a PKPD-model describing the time course of QT prolongation for citalopram To evaluate the effect of charcoal administration on the relative risk of TdP after citalopram overdose

7 Data set 53 patients who had taken citalopram in an overdose event 63 events 36 females (68%) years (median 30 years) Reported dose: mg (1 85 tablets) Single dose activated charcoal on 17 events ( hours after the overdose) 39 patients were taking citalopram therapeutically No case of TdP

8 Veracity grade of dosing history Veracity grade Description Excellent history Good history Less reliable history Poor history Very poor history # of events

9 Dose-normalised (30 mg) concentrations vs. time (n=189) Dose-normalised concentration (mg/l) No citalopram therapeutically No charcoal Citalopram therapeutically No charcoal Time (hours) Citalopram therapeutically Charcoal

10 Observed QT-RR intervals 167 QT and RR combinations 33 combinations at increased risk bpm 60 bpm QT Interval (ms) RR Interval (ms)

11 Data analysis WinBUGS v. 1.4 Prior information PK Informative 14 PK studies on citalopram taken in therapeutic doses PD Low-information Biologically plausible

12 Final PK model Uncertainty in dose ~ veracity grade + F Ka 1-compartment CL Uncertainty X in time of ingestion Reduced F by 22% Activated charcoal Increased CL by 72%

13 Dose-dependent clearance? Interactions with co-ingestants? Metabolised by CYP3A4 and/or CYP2C19 CL/F (L/h) 20 CL/F (L/h) Dose (mg) 0 No co-ingestant Co-ingestant(s)

14 Simulation study Bias in PK parameters in 30 data sets with the same design 100 Added uncertainty in dose 50 0 Bias (%) No uncertainty in dose CL Ka Ω CL Ω Ka f CL-charcoal Vd Conc Base Ω Vd Ω CBase f F-charcoal

15 PKPD model of QT interval prolongation QT ij = QTc ij RR αi ij QTc ij = QTc i,0 + Slope i Ce ij + QTc i, co ingestant drugs t eq = 1.4 h α = 0.36 Slope = 40 L ms/mgl QTc QTc i,0 9 ms higher in women increased with age QTc i, < 5 ms i,co-ingestant drugs

16 Simulation of probability for risk of TdP from PKPD model 2000 patients 10 dose levels: 5-90 x DDD (Defined Daily Dose; 20 mg) Median of observed RR intervals (760 ms) QT interval with increased risk of TdP (447 ms)

17 Probability of risk for TdP with and without activated charcoal RR = 80 bpm (= 760 ms ) Without charcoal With charcoal

18 Relative reduction in hazard/risk for TdP by charcoal

19 Conclusions Informative priors and veracity judgements on dosing history could be used for developing a PKPD model from overdose data Activated charcoal reduced F and increased CL after citalopram overdoses QT prolongation was delayed relative to C max of citalopram Administration of activated charcoal reduced the risk for TdP by approximately 60%

20 Acknowledgements Staff of the Clinical Toxicology and Emergency Departments, Newcastle Mater Hospitals, Newcastle, Australia Knut & Alice Wallenberg Foundation, Sweden